Empowering Optimal Control with Machine Learning: A Perspective from Model Predictive Control

TL;DR

This paper reviews how machine learning enhances model predictive control in solving complex optimal control problems, highlighting recent progress and discussing challenges in robustness and implementation.

Contribution

It provides a comprehensive survey of recent machine learning applications in model predictive control, emphasizing new methods and challenges.

Findings

Machine learning improves the efficiency of optimal control solvers.

Recent advances enable more robust and adaptable control algorithms.

Challenges include ensuring reliability and interpretability of ML-enhanced control systems.

Abstract

Solving complex optimal control problems have confronted computational challenges for a long time. Recent advances in machine learning have provided us with new opportunities to address these challenges. This paper takes model predictive control, a popular optimal control method, as the primary example to survey recent progress that leverages machine learning techniques to empower optimal control solvers. We also discuss some of the main challenges encountered when applying machine learning to develop more robust optimal control algorithms.